All language subtitles for [English (auto-generated)] PSY 294_ G_Power tutorial (t-tests) [DownSub.com]

this is the screencast tutorial for how

to use the program g-power to determine

two things first of all is a priori

power that is computing the required

sample size that you need to achieve a

desired level of power for a study given

levels of alpha and effect size and the

other thing is to compute post hoc power

that is exactly what is the obtained

level of power you have in the study

given your sample statistics to estimate

things like effect size as well as the

Alpha level that you use in your study

now the nice thing about G power is as

you can see here there are a variety of

statistical tests that you can select

from so we're not going to cover all of

these this semester or last semester or

even over the course of this curriculum

okay but there really is a lot of

flexibility in the program G power for

any design that you might come up with

there's different test families as well

right now we're going to stick with just

the T tests but as you can see F tests

which may be required for your upcoming

project are available in here as well in

addition to things like chi-square and D

tests that we've done in the past again

what we're going to do to stick with T

tests for now and let's look at the

first example say that what we want to

do is to determine what sample size will

be required to achieve a desired level

of power for a paired samples t-test so

we know we're in the t-test family and

then we go over to statistical tests and

find the test in which we're interested

in this case a paired samples are what

they're calling a matched pairs t-test

is indeed what we're looking at the

difference between two dependent means

so we select that test now as I

mentioned in G power all you have to do

is to fill in the different values that

you know and it will calculate the

unknown value for you so in this case

the first thing we want to do is to tell

it what type of tests that we're looking

at a priori or pose talk now you can see

there are a lot of other options in here

as well but these are going to be the

only two in which we're interested for

now let's do the a priori test which is

going to tell us what is our required

sample size to achieve a desired level

of power then we need to do is to let it

know whether it's a directional or a non

directional hypothesis by telling it the

number of tails we know that a non

directional test

a two-tailed test a directional test is

a one-tailed test let's assume for now

that we're going to going with the

two-tailed test now as mentioned in

lecture all we then to know are what are

our effect size our alpha level and our

desired power and then what we can do is

to then compute the sample size that

would be necessary in order to achieve

that level of power so the effect size

here at 0.5 is a moderate effect size we

can simply change that by 2 anything we

want if we think it'll be a little bit

larger 0.6 if we think it would be a

small effect size of 0.2 0.8 for a large

effect size those are conventional

levels so let's just stick with a

moderate effect size of 0.5 our alpha

level is indeed going to be a value of

0.05 as it is always going to be in this

course and in behavioral science in

general and then what we can do is to

say what is the level of power that we

would desire now point nine five is a

great level of power recall what this

means is that if there really is an

effect we want there to be a 95% chance

of detecting this effect in our study

now we might not want such a strange

requirement let's lower this down to

something like 90% and that's all it

takes is inputting your effect size

estimate D your alpha level and your

desired power and then clicking on

calculate now what it's going to produce

is a graph similar to the one that we

saw in lecture as well ok where it's

showing you the red distribution

represents the null distribution the

dashed blue distribution represents the

potential alternative research

hypothesis and you can see the shaded

areas for alpha and beta similar to the

way they were discussed in lecture now

what's going to be important for us then

is looking at what is going to be the

sample size that's required to achieve

this level of power the place you can

find that is right here under total

sample size and we can see that we need

a total of 44 participants in this study

that is in a within subject situation

where we're comparing two dependent

means we would need forty four people

and then of course we know this would be

within subject study where these 44

people are participating in both

conditions now there's a couple other

things we want to look at

here in particular one thing we can do

is to look at the XY plot for a range of

possible values if we want to do that

then all you need to do is to click on

the button at the bottom that brings up

this plotting window and you can change

everything you want in here as well you

can change the effect size you can

change your alpha level and you can set

some other specific details as well for

example say that what you want to see is

the sample size required to achieve

different levels of power ranging from

0.5 all the way up to 0.95 once you set

the details in the values how you like

them then all you have to do is click

draw plot and then it will show you the

values specifically to achieve a certain

level of power which is shown on the

x-axis say that we want to achieve power

of 0.8 we can use this plot then to see

how many participants are going to be

required if we read over on the y-axis

it's going to show us about 33

participants are required total sample

size given our effect size and alpha

level in order to achieve that power of

0.8 in a trailer that simple there's a

second example we can look at as well so

say that you don't have your effect size

in particular for example what you may

be doing is basing your effect size off

of previous research so you run one

study things didn't go so well

yet you have the data the sample data

from that previous study that you can

use to estimate effect size for an

upcoming study for example any situation

like this where you want to use the

sample statistics to determine your

effect size can be done by clicking on

this determined button here on the left

what that's going to do is to open up

this little side window or drawer and in

here what you can do is to calculate the

sample statistics enter them in directly

and then it will determine the effect

size for you okay so for example we know

that when we're calculating a paired

samples t-test we end up calculating a

mean different score and a standard

deviation of that different score so if

you come and select this radio button

for from differences and say that we've

done a study where we find

mean difference of say two and a

standard deviation among different

scores of say five point five then we

can click this calculate button here at

the bottom and it will do the effect

size calculation for us now this is a

pretty simple calculation which I'm sure

you can verify using a calculator as

well but if you have these sample

statistics handy this is a convenient

way for it to calculate the effect size

and then by clicking the other button

transfer to main window then you can see

what it does is it copies the effect

size into the main window of g-power

directly for you from there everything

else perceived as normal you can simply

click calculate to determine your total

sample size necessary in this case it

would be 82 you can see as the effect

size went down from 0.5 to 0.36 the

sample size required went up to maintain

the same level of power and once again

you can click for the XY plot to show an

entire range of values so let's look at

one other type of test the independent

samples t-test because although the

logic is the same and I'm sure you could

figure it out for yourself but it is

handled a little bit differently so

let's take a look at that one now as

well again this is still going to be a

t-test but specifically this is going to

be differences between means of two

independent groups this then is the

independent samples t-test in g-power so

clicking on here again what we may still

want to know is what is the sample size

that's going to be required to achieve a

specific desired level of power so again

everything is the same so let's look at

the same effect size of 0.5 alpha of

0.05 power again and let's say 0.9 0

just changing the values in the boxes

here okay the only other box that's on

here is the allocation ratio n2 to n1

now specifically what we're going to

strive for in an independent samples

test is to have equal sample sizes in

the two groups well if that's the case

in the ratio of sample size between the

groups is going to be equal to one but

it may not always be the case for now

let's assume it is and if it is you can

simply click calculate and then it's

going to show you the sample size in

each group 70 instead

for a total sample size of 140 now if

you have a different allocation ratio

let's say that you have different

numbers of people within the two

different groups say that it's an

allocation issue of 0.75 okay this would

be an example if you have say 30 people

in one group and 40 people in the other

group well then the number of people in

one group 30 over the number of people

in the other group 40 would equal 3/4 or

0.75 in this case you still just click

calculate and again it's going to show

you the sample size that you need which

you can see is student 42 here and then

how they're going to be allocated across

the two different groups now you may

notice that the total sample size

necessary in this case has gone up a

little bit okay this is an important

point in an independent samples t-test

that you actually achieve the highest

level of power if your sample your total

sample is allocated evenly across the

two groups now another thing that we can

do here is to think about using the

determined window in order to enter the

sample statistics just like we did for

the paired samples test okay now in this

case again what we may want to do is

just enter the sample data say we've

collected some data we find the mean of

the first group is 55 mean of the second

group is 45 Center deviation 1 could be

6 say the other group is 7 we can use

this to calculate our effect size here

as well transfer it to the main window

which it's done and then complete the

calculation for us okay in this case of

course with such a large effect size we

see that we need a relatively fewer

people in the two groups now if we don't

have equal effect sizes which is the top

radio button here in this calculation or

determine drawer then what we may do is

to enter in the mean of the two groups

again let's say it's 55 and 45 then what

it asks for here is the standard

deviation within each group now it's

important to know what this is going to

be is essentially the pooled standard

deviation that is the school

a root of the pooled variance estimate

that we know how to calculate let's say

that's something like six point two just

to select a number again we can

calculate the effect size transfer it to

the main window and then calculate the

sample size necessary to produce the

desired level of power once again with

such a large effect size the sample size

of a needing each group is relatively

pretty small so finally let's look at an

example of how we can then calculate

post-talk power now remember this is

going to change a little bit because

what we're looking at finding now is not

for a given level of power what is our

sample size but given our sample size

what is our achieved level of power now

for this we're just going to stick with

independent samples t-test and doing it

for the dependent test is very similar

so I'm not going to go back through the

motions there but again where you can

see it has happened to retain the values

that we calculated in our last analysis

this may or may not be the case okay but

what you can do in this case especially

is because you're doing this post hoc

you're typically going to have exactly

the values that would go in these boxes

over here okay so just to change the

numbers to produce a different example

say we have one class that get 275

average on an exam another class get 272

average we have the standard deviation

for the exam scores in each group if

that's the case again we can just

calculate the effect size so these are

going to be values that are coming

directly out of your analysis either

through SPSS or Excel or wherever you're

going to calculate them once again all

we have to do is transfer to the main

window okay and now then we're also

going to know the sample we had in each

group let's say that we had 18 people in

the first group and 20 people in the

second group now in this case when we

calculate it's going to show us is the

calculated the achieved level of power

in this case rather low it's 0.27 and

you can see that in terms of how we

talked about it in lecture especially

with the overlap here of the alternative

distribution and specifically how much

that is

the left of our critical value which is

shown in green well that's the last

example that I wanted to go through what

we've seen now is how to do two things

how to calculate a priority power that

is for a given desired level of power

what is the sample size necessary to

achieve it and post-hoc power given all

of our sample statistics that is the

values that are associated with our

exact experimental design or our study

that is our sample means in standard

deviations as well as sample size what

is the obtained level of power that

we've achieved within our study this

should be enough to prepare you to not

only complete the homework given the

effect sizes and other calculations as

well as the values that are provided in

the homework problems themselves but

it's also going to start to familiarize

you with G power so that you can do a

power analysis for your own upcoming

project now again the basic logic is the

same even though exactly the boxes you

click on are going to be slightly

different for your study depending on

the exact nature of the design but

that's something your lab instructor is

going to walk you through as well

depending on whether or not for example

you have a factorial design or situation

we have a single independent variable

with multiple levels so forth and so on

so thanks for tuning in today and good

luck with Alan work